---
name: al-504-build-your-own-support-chat-agent-instead-of-paying-for-a
description: "Jenny Rudd runs Dispute Buddy, an app that helps people find patterns of coercive control in their texts. In a LinkedIn comment in October 2026 she said they saved $3,600 a year by replacing their paid off the shelf support chat with a customer service agent they built themselves. She says it diagnoses and fixes problems customers hit in the app and talks to them in her voice, knowing they may be going through a hard time, and that it is 100 times better than the old one. Self-reported in a comment, with no setup details. When customers can be in a hard place, make reaching a person easy. Use when the user wants help with support chat agent (customer support) for a small business."
---

# Build your own support chat agent instead of paying for an off the shelf one: teach it your top 10 customer problems, your fixes and your tone, test it on past questions, and have it hand upset customers to you

Jenny Rudd runs Dispute Buddy, an app that helps people find patterns of coercive control in their texts. In a LinkedIn comment in October 2026 she said they saved $3,600 a year by replacing their paid off the shelf support chat with a customer service agent they built themselves. She says it diagnoses and fixes problems customers hit in the app and talks to them in her voice, knowing they may be going through a hard time, and that it is 100 times better than the old one. Self-reported in a comment, with no setup details. When customers can be in a hard place, make reaching a person easy.

## Inputs to collect first
Ask the user for anything below that they haven't already given you. Don't guess missing facts.

- Cost
- 3 months
- Product
- Safety, money, legal or sensitive topics
- Name/email
- Week

## Instructions

Do the task below with the user's inputs in place of each [bracket].

1. Check what your current support chat costs a year: [cost]. Export the last [3 months] of chats or emails.
2. Ask AI: "From these conversations, list our 10 most common customer problems and how often each comes up."
3. Write the fix for each problem in plain steps, and paste 5 of your own replies so it can learn your tone.
4. Say: "Act as our support agent for [product]. Answer the way I would, walk the customer through the fix step by step, and keep replies short and kind."
5. Add handoff rules: "If the customer is upset, mentions [safety, money, legal or sensitive topics], or the fix doesn't work after two tries, stop and hand the chat to [name/email] with a summary."
6. Test it on 20 real past questions before going live, and note every wrong or made-up answer.
7. Fix the instructions, go live, and review a sample of chats every [week].
Safety: never let the agent invent policies, refunds or promises, and make a human easy to reach at every step.

## Safety checks (share these with the user)

- Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes.
- Anything legal, tax or compliance-related is a first draft for a qualified person to check. It is not professional advice, and clause references and deadlines must be checked against the source.
- Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings.
- Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically".
- Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed.

## About this skill

- Works with: ChatGPT, Claude, Intercom, Zendesk
- Based on this real-world example: https://aibusinessplays.com/plays/app-founder-builds-own-customer-service-agent-saves-3600-a-year/
- Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-504-build-your-own-support-chat-agent-instead-of-paying-for-an-o/
- To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/